Convolutional Neural Networks

Project: Write an Algorithm for a Dog Identification App


In this notebook, some template code has already been provided for you, and you will need to implement additional functionality to successfully complete this project. You will not need to modify the included code beyond what is requested. Sections that begin with '(IMPLEMENTATION)' in the header indicate that the following block of code will require additional functionality which you must provide. Instructions will be provided for each section, and the specifics of the implementation are marked in the code block with a 'TODO' statement. Please be sure to read the instructions carefully!

Note: Once you have completed all of the code implementations, you need to finalize your work by exporting the Jupyter Notebook as an HTML document. Before exporting the notebook to html, all of the code cells need to have been run so that reviewers can see the final implementation and output. You can then export the notebook by using the menu above and navigating to File -> Download as -> HTML (.html). Include the finished document along with this notebook as your submission.

In addition to implementing code, there will be questions that you must answer which relate to the project and your implementation. Each section where you will answer a question is preceded by a 'Question X' header. Carefully read each question and provide thorough answers in the following text boxes that begin with 'Answer:'. Your project submission will be evaluated based on your answers to each of the questions and the implementation you provide.

Note: Code and Markdown cells can be executed using the Shift + Enter keyboard shortcut. Markdown cells can be edited by double-clicking the cell to enter edit mode.

The rubric contains optional "Stand Out Suggestions" for enhancing the project beyond the minimum requirements. If you decide to pursue the "Stand Out Suggestions", you should include the code in this Jupyter notebook.


Why We're Here

In this notebook, you will make the first steps towards developing an algorithm that could be used as part of a mobile or web app. At the end of this project, your code will accept any user-supplied image as input. If a dog is detected in the image, it will provide an estimate of the dog's breed. If a human is detected, it will provide an estimate of the dog breed that is most resembling. The image below displays potential sample output of your finished project (... but we expect that each student's algorithm will behave differently!).

Sample Dog Output

In this real-world setting, you will need to piece together a series of models to perform different tasks; for instance, the algorithm that detects humans in an image will be different from the CNN that infers dog breed. There are many points of possible failure, and no perfect algorithm exists. Your imperfect solution will nonetheless create a fun user experience!

The Road Ahead

We break the notebook into separate steps. Feel free to use the links below to navigate the notebook.

  • Step 0: Import Datasets
  • Step 1: Detect Humans
  • Step 2: Detect Dogs
  • Step 3: Create a CNN to Classify Dog Breeds (from Scratch)
  • Step 4: Create a CNN to Classify Dog Breeds (using Transfer Learning)
  • Step 5: Write your Algorithm
  • Step 6: Test Your Algorithm

Step 0: Import Datasets

Make sure that you've downloaded the required human and dog datasets:

  • Download the dog dataset. Unzip the folder and place it in this project's home directory, at the location /dogImages.

  • Download the human dataset. Unzip the folder and place it in the home directory, at location /lfw.

Note: If you are using a Windows machine, you are encouraged to use 7zip to extract the folder.

In the code cell below, we save the file paths for both the human (LFW) dataset and dog dataset in the numpy arrays human_files and dog_files.

In [1]:
import os
import random
import requests
import time
import ast
import numpy as np
from glob import glob
import cv2                
from tqdm import tqdm
from PIL import Image, ImageFile 

import torch
import torchvision
from torchvision import datasets
import torchvision.transforms as transforms
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim
import torchvision.models as models

import matplotlib.pyplot as plt                        
%matplotlib inline

ImageFile.LOAD_TRUNCATED_IMAGES = True

# check if CUDA is available
use_cuda = torch.cuda.is_available()
In [2]:
import numpy as np
from glob import glob

# load filenames for human and dog images
human_files = np.array(glob("data/lfw/*/*"))
dog_files = np.array(glob("data/dogImages/*/*/*"))

# print number of images in each dataset
print('There are %d total human images.' % len(human_files))
print('There are %d total dog images.' % len(dog_files))
There are 13233 total human images.
There are 8351 total dog images.

Step 1: Detect Humans

In this section, we use OpenCV's implementation of Haar feature-based cascade classifiers to detect human faces in images.

OpenCV provides many pre-trained face detectors, stored as XML files on github. We have downloaded one of these detectors and stored it in the haarcascades directory. In the next code cell, we demonstrate how to use this detector to find human faces in a sample image.

In [3]:
import cv2                
import matplotlib.pyplot as plt                        
%matplotlib inline   


def convertRGB(img): 
    return cv2.cvtColor(img, cv2.COLOR_BGR2RGB)

# extract pre-trained face detector
face_cascade = cv2.CascadeClassifier('haarcascades/haarcascade_frontalface_alt.xml')

# load color (BGR) image
img = cv2.imread(human_files[10])
# convert BGR image to grayscale
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)

# find faces in image
faces = face_cascade.detectMultiScale(gray)

# print number of faces detected in the image
print('Number of faces detected:', len(faces))

# get bounding box for each detected face
for (x,y,w,h) in faces:
    # add bounding box to color image
    cv2.rectangle(img,(x,y),(x+w,y+h),(255,0,0),2)
    
# convert BGR image to RGB for plotting
cv_rgb = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)

# display the image, along with bounding box
plt.imshow(cv_rgb)
plt.show()
Number of faces detected: 1

Before using any of the face detectors, it is standard procedure to convert the images to grayscale. The detectMultiScale function executes the classifier stored in face_cascade and takes the grayscale image as a parameter.

In the above code, faces is a numpy array of detected faces, where each row corresponds to a detected face. Each detected face is a 1D array with four entries that specifies the bounding box of the detected face. The first two entries in the array (extracted in the above code as x and y) specify the horizontal and vertical positions of the top left corner of the bounding box. The last two entries in the array (extracted here as w and h) specify the width and height of the box.

Write a Human Face Detector

We can use this procedure to write a function that returns True if a human face is detected in an image and False otherwise. This function, aptly named face_detector, takes a string-valued file path to an image as input and appears in the code block below.

In [4]:
# returns "True" if face is detected in image stored at img_path
def face_detector(img_path):
    img = cv2.imread(img_path)
    gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
    faces = face_cascade.detectMultiScale(gray)
    return len(faces) > 0

(IMPLEMENTATION) Assess the Human Face Detector

Question 1: Use the code cell below to test the performance of the face_detector function.

  • What percentage of the first 100 images in human_files have a detected human face?
  • What percentage of the first 100 images in dog_files have a detected human face?

Ideally, we would like 100% of human images with a detected face and 0% of dog images with a detected face. You will see that our algorithm falls short of this goal, but still gives acceptable performance. We extract the file paths for the first 100 images from each of the datasets and store them in the numpy arrays human_files_short and dog_files_short.

Answer: (You can print out your results and/or write your percentages in this cell)

In [5]:
from tqdm import tqdm

human_files_short = human_files[:100]
dog_files_short = dog_files[:100]

#-#-# Do NOT modify the code above this line. #-#-#

## TODO: Test the performance of the face_detector algorithm 
## on the images in human_files_short and dog_files_short.

detected_faces_in_humans = 0
detected_faces_in_dogs = 0

for i in range(100):
    if face_detector(human_files_short[i]):
        detected_faces_in_humans += 1
    if face_detector(dog_files_short[i]):
        detected_faces_in_dogs +=1
        
print (f"Detected human faces: {detected_faces_in_humans}%")
print (f"Detected faces in dogs: {detected_faces_in_dogs}%")
Detected human faces: 96%
Detected faces in dogs: 18%
In [6]:
### (Optional) 
### TODO: Test performance of anotherface detection algorithm.
### Feel free to use as many code cells as needed.
#lbp_face_cascade = cv2.CascadeClassifier('lbpcascades/lbpcascade_frontalface.xml')  
lbpcascades_face_cascade = cv2.CascadeClassifier('lbpcascades/lbpcascade_frontalface.xml') 


# print number of faces detected in the image
print('Number of faces detected:', len(faces))
    
def lbpcascades_face_detector(img_path):
    img = cv2.imread(img_path)
    gray = convertRGB(img)
    faces = lbpcascades_face_cascade.detectMultiScale(gray)
    return len(faces) > 0
Number of faces detected: 1
In [7]:
detected_faces_in_humans = 0
detected_faces_in_dogs = 0

for i in range(100):
    if lbpcascades_face_detector(human_files_short[i]):
        detected_faces_in_humans += 1
    if lbpcascades_face_detector(dog_files_short[i]):
        detected_faces_in_dogs +=1
        
print (f"Detected human faces using lbpcascades: {detected_faces_in_humans}%")
print (f"Detected faces in dogsusing lbpcascades: {detected_faces_in_dogs}%")
Detected human faces using lbpcascades: 89%
Detected faces in dogsusing lbpcascades: 26%

We suggest the face detector from OpenCV as a potential way to detect human images in your algorithm, but you are free to explore other approaches, especially approaches that make use of deep learning :). Please use the code cell below to design and test your own face detection algorithm. If you decide to pursue this optional task, report performance on human_files_short and dog_files_short.


Step 2: Detect Dogs

In this section, we use a pre-trained model to detect dogs in images.

Obtain Pre-trained VGG-16 Model

The code cell below downloads the VGG-16 model, along with weights that have been trained on ImageNet, a very large, very popular dataset used for image classification and other vision tasks. ImageNet contains over 10 million URLs, each linking to an image containing an object from one of 1000 categories.

In [8]:
import torch
import torchvision.models as models

# define VGG16 model
VGG16 = models.vgg16(pretrained=True)

# check if CUDA is available
use_cuda = torch.cuda.is_available()

# move model to GPU if CUDA is available
if use_cuda:
    VGG16 = VGG16.cuda()
In [9]:
VGG16
Out[9]:
VGG(
  (features): Sequential(
    (0): Conv2d(3, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
    (1): ReLU(inplace)
    (2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
    (3): ReLU(inplace)
    (4): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
    (5): Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
    (6): ReLU(inplace)
    (7): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
    (8): ReLU(inplace)
    (9): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
    (10): Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
    (11): ReLU(inplace)
    (12): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
    (13): ReLU(inplace)
    (14): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
    (15): ReLU(inplace)
    (16): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
    (17): Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
    (18): ReLU(inplace)
    (19): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
    (20): ReLU(inplace)
    (21): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
    (22): ReLU(inplace)
    (23): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
    (24): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
    (25): ReLU(inplace)
    (26): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
    (27): ReLU(inplace)
    (28): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
    (29): ReLU(inplace)
    (30): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
  )
  (avgpool): AdaptiveAvgPool2d(output_size=(7, 7))
  (classifier): Sequential(
    (0): Linear(in_features=25088, out_features=4096, bias=True)
    (1): ReLU(inplace)
    (2): Dropout(p=0.5)
    (3): Linear(in_features=4096, out_features=4096, bias=True)
    (4): ReLU(inplace)
    (5): Dropout(p=0.5)
    (6): Linear(in_features=4096, out_features=1000, bias=True)
  )
)

Given an image, this pre-trained VGG-16 model returns a prediction (derived from the 1000 possible categories in ImageNet) for the object that is contained in the image.

(IMPLEMENTATION) Making Predictions with a Pre-trained Model

In the next code cell, you will write a function that accepts a path to an image (such as 'dogImages/train/001.Affenpinscher/Affenpinscher_00001.jpg') as input and returns the index corresponding to the ImageNet class that is predicted by the pre-trained VGG-16 model. The output should always be an integer between 0 and 999, inclusive.

Before writing the function, make sure that you take the time to learn how to appropriately pre-process tensors for pre-trained models in the PyTorch documentation.

In [10]:
def image_to_tensor(img_path):
    '''
    As per Pytorch documentations: All pre-trained models expect input images normalized in the same way, 
    i.e. mini-batches of 3-channel RGB images
    of shape (3 x H x W), where H and W are expected to be at least 224. 
    The images have to be loaded in to a range of [0, 1] and 
    then normalized using mean = [0.485, 0.456, 0.406] and std = [0.229, 0.224, 0.225]. 
    You can use the following transform to normalize:
    '''
    img = Image.open(img_path).convert('RGB')
    transformations = transforms.Compose([transforms.Resize(size=224),
                                          transforms.CenterCrop((224,224)),
                                         transforms.ToTensor(),
                                         transforms.Normalize(mean=[0.485, 0.456, 0.406],
                                                              std=[0.229, 0.224, 0.225])])
    image_tensor = transformations(img)[:3,:,:].unsqueeze(0)
    return image_tensor



def im_convert(tensor):
    """ Display a tensor as an image. """
    
    image = tensor.to("cpu").clone().detach()
    image = image.numpy().squeeze()
    image = image.transpose(1,2,0)
    image = image * np.array((0.229, 0.224, 0.225)) + np.array((0.485, 0.456, 0.406))
    image = image.clip(0, 1)

    return image
In [11]:
dog_image = Image.open('data/dogImages/train/001.Affenpinscher/Affenpinscher_00001.jpg')
plt.imshow(dog_image)
plt.show()
In [12]:
tensordog = image_to_tensor('data/dogImages/train/001.Affenpinscher/Affenpinscher_00001.jpg')
# print(test_tensor)
print(tensordog.shape)
plt.imshow(im_convert(tensordog))
torch.Size([1, 3, 224, 224])
Out[12]:
<matplotlib.image.AxesImage at 0x14d097a6e80>
In [13]:
from PIL import Image
import torchvision.transforms as transforms

# Set PIL to be tolerant of image files that are truncated.
from PIL import ImageFile
ImageFile.LOAD_TRUNCATED_IMAGES = True

def VGG16_predict(img_path):
    '''
    Use pre-trained VGG-16 model to obtain index corresponding to 
    predicted ImageNet class for image at specified path
    
    Args:
        img_path: path to an image
        
    Returns:
        Index corresponding to VGG-16 model's prediction
        
        
        
    '''
    
     ## TODO: Complete the function.
    ## Load and pre-process an image from the given img_path
    ## Return the *index* of the predicted class for that image
    image_tensor = image_to_tensor(img_path)
   
    # move model inputs to cuda, if GPU available
    if use_cuda:
        image_tensor = image_tensor.cuda()

    # get sample outputs
    output = VGG16(image_tensor)
    # convert output probabilities to predicted class
    _, preds_tensor = torch.max(output, 1)
    pred = np.squeeze(preds_tensor.numpy()) if not use_cuda else np.squeeze(preds_tensor.cpu().numpy())
 
    return int(pred) # predicted class index
    
   
    
   
In [14]:
import requests
import ast

LABELS_MAP_URL = "https://gist.githubusercontent.com/yrevar/942d3a0ac09ec9e5eb3a/raw/c2c91c8e767d04621020c30ed31192724b863041/imagenet1000_clsid_to_human.txt"

def get_human_readable_label_for_class_id(class_id):
    labels = ast.literal_eval(requests.get(LABELS_MAP_URL).text)
    print(f"Label:{labels[class_id]}")
    return labels[class_id]
    
    
test_prediction = VGG16_predict('data/dogImages/train/001.Affenpinscher/Affenpinscher_00001.jpg')
pred_class = int(test_prediction)

print(f"Predicted class id: {pred_class}")
class_description = get_human_readable_label_for_class_id(pred_class)
print(f"Predicted class for image is *** {class_description.upper()} ***")
Predicted class id: 252
Label:affenpinscher, monkey pinscher, monkey dog
Predicted class for image is *** AFFENPINSCHER, MONKEY PINSCHER, MONKEY DOG ***

(IMPLEMENTATION) Write a Dog Detector

While looking at the dictionary, you will notice that the categories corresponding to dogs appear in an uninterrupted sequence and correspond to dictionary keys 151-268, inclusive, to include all categories from 'Chihuahua' to 'Mexican hairless'. Thus, in order to check to see if an image is predicted to contain a dog by the pre-trained VGG-16 model, we need only check if the pre-trained model predicts an index between 151 and 268 (inclusive).

Use these ideas to complete the dog_detector function below, which returns True if a dog is detected in an image (and False if not).

In [15]:
### returns "True" if a dog is detected in the image stored at img_path
def dog_detector(img_path):
    ## TODO: Complete the function.
    prediction = VGG16_predict(img_path)
    
    return ((prediction >= 151) & (prediction <=268))

(IMPLEMENTATION) Assess the Dog Detector

Question 2: Use the code cell below to test the performance of your dog_detector function.

  • What percentage of the images in human_files_short have a detected dog?
  • What percentage of the images in dog_files_short have a detected dog?

Answer:

In [16]:
### TODO: Test the performance of the dog_detector function
### on the images in human_files_short and dog_files_short.
detected_dogs_in_humans = 0
detected_dogs_in_dogs = 0

for j in range(100):
    if dog_detector(human_files_short[j]):
        detected_dogs_in_humans += 1
        print(f"This human ({j}) looks like a dog")
        human_dog_image = Image.open(human_files_short[j])
        plt.imshow(human_dog_image)
        plt.show()
    if dog_detector(dog_files_short[j]):
        detected_dogs_in_dogs +=1
        
print (f"Percentage of the images in human_files_short that have a detected dog: {detected_dogs_in_humans}%")
print (f"Percentage of the images in dog_files_short that have a detected dog: {detected_dogs_in_dogs}%")
Percentage of the images in human_files_short that have a detected dog: 0%
Percentage of the images in dog_files_short that have a detected dog: 93%

We suggest VGG-16 as a potential network to detect dog images in your algorithm, but you are free to explore other pre-trained networks (such as Inception-v3, ResNet-50, etc). Please use the code cell below to test other pre-trained PyTorch models. If you decide to pursue this optional task, report performance on human_files_short and dog_files_short.

In [17]:
### (Optional) 
### TODO: Report the performance of another pre-trained network.
### Feel free to use as many code cells as needed.

Step 3: Create a CNN to Classify Dog Breeds (from Scratch)

Now that we have functions for detecting humans and dogs in images, we need a way to predict breed from images. In this step, you will create a CNN that classifies dog breeds. You must create your CNN from scratch (so, you can't use transfer learning yet!), and you must attain a test accuracy of at least 10%. In Step 4 of this notebook, you will have the opportunity to use transfer learning to create a CNN that attains greatly improved accuracy.

We mention that the task of assigning breed to dogs from images is considered exceptionally challenging. To see why, consider that even a human would have trouble distinguishing between a Brittany and a Welsh Springer Spaniel.

Brittany Welsh Springer Spaniel

It is not difficult to find other dog breed pairs with minimal inter-class variation (for instance, Curly-Coated Retrievers and American Water Spaniels).

Curly-Coated Retriever American Water Spaniel

Likewise, recall that labradors come in yellow, chocolate, and black. Your vision-based algorithm will have to conquer this high intra-class variation to determine how to classify all of these different shades as the same breed.

Yellow Labrador Chocolate Labrador Black Labrador

We also mention that random chance presents an exceptionally low bar: setting aside the fact that the classes are slightly imabalanced, a random guess will provide a correct answer roughly 1 in 133 times, which corresponds to an accuracy of less than 1%.

Remember that the practice is far ahead of the theory in deep learning. Experiment with many different architectures, and trust your intuition. And, of course, have fun!

(IMPLEMENTATION) Specify Data Loaders for the Dog Dataset

Use the code cell below to write three separate data loaders for the training, validation, and test datasets of dog images (located at dogImages/train, dogImages/valid, and dogImages/test, respectively). You may find this documentation on custom datasets to be a useful resource. If you are interested in augmenting your training and/or validation data, check out the wide variety of transforms!

In [18]:
import os
from torchvision import datasets

### TODO: Write data loaders for training, validation, and test sets
## Specify appropriate transforms, and batch_sizes
batch_size = 16

# number of subprocesses to use for data loading
num_workers = 2

# convert data to a normalized torch.FloatTensor
transform = transforms.Compose([transforms.Resize(size=224),
                                transforms.CenterCrop((224,224)),
                                transforms.RandomHorizontalFlip(), # randomly flip and rotate
                                transforms.RandomRotation(10),
                                transforms.ToTensor(),
                                transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])])

# define training, test and validation data directories
data_dir = 'data/dogImages/'

image_datasets = {x: datasets.ImageFolder(os.path.join(data_dir, x), transform)
                  for x in ['train', 'valid', 'test']}
loaders_scratch = {
    x: torch.utils.data.DataLoader(image_datasets[x], shuffle=True, batch_size=batch_size, num_workers=num_workers)
    for x in ['train', 'valid', 'test']}
In [19]:
class_names = image_datasets['train'].classes
nb_classes = len(class_names)

print("Number of classes:", nb_classes)
print("\nClass names: \n\n", class_names)
Number of classes: 133

Class names: 

 ['001.Affenpinscher', '002.Afghan_hound', '003.Airedale_terrier', '004.Akita', '005.Alaskan_malamute', '006.American_eskimo_dog', '007.American_foxhound', '008.American_staffordshire_terrier', '009.American_water_spaniel', '010.Anatolian_shepherd_dog', '011.Australian_cattle_dog', '012.Australian_shepherd', '013.Australian_terrier', '014.Basenji', '015.Basset_hound', '016.Beagle', '017.Bearded_collie', '018.Beauceron', '019.Bedlington_terrier', '020.Belgian_malinois', '021.Belgian_sheepdog', '022.Belgian_tervuren', '023.Bernese_mountain_dog', '024.Bichon_frise', '025.Black_and_tan_coonhound', '026.Black_russian_terrier', '027.Bloodhound', '028.Bluetick_coonhound', '029.Border_collie', '030.Border_terrier', '031.Borzoi', '032.Boston_terrier', '033.Bouvier_des_flandres', '034.Boxer', '035.Boykin_spaniel', '036.Briard', '037.Brittany', '038.Brussels_griffon', '039.Bull_terrier', '040.Bulldog', '041.Bullmastiff', '042.Cairn_terrier', '043.Canaan_dog', '044.Cane_corso', '045.Cardigan_welsh_corgi', '046.Cavalier_king_charles_spaniel', '047.Chesapeake_bay_retriever', '048.Chihuahua', '049.Chinese_crested', '050.Chinese_shar-pei', '051.Chow_chow', '052.Clumber_spaniel', '053.Cocker_spaniel', '054.Collie', '055.Curly-coated_retriever', '056.Dachshund', '057.Dalmatian', '058.Dandie_dinmont_terrier', '059.Doberman_pinscher', '060.Dogue_de_bordeaux', '061.English_cocker_spaniel', '062.English_setter', '063.English_springer_spaniel', '064.English_toy_spaniel', '065.Entlebucher_mountain_dog', '066.Field_spaniel', '067.Finnish_spitz', '068.Flat-coated_retriever', '069.French_bulldog', '070.German_pinscher', '071.German_shepherd_dog', '072.German_shorthaired_pointer', '073.German_wirehaired_pointer', '074.Giant_schnauzer', '075.Glen_of_imaal_terrier', '076.Golden_retriever', '077.Gordon_setter', '078.Great_dane', '079.Great_pyrenees', '080.Greater_swiss_mountain_dog', '081.Greyhound', '082.Havanese', '083.Ibizan_hound', '084.Icelandic_sheepdog', '085.Irish_red_and_white_setter', '086.Irish_setter', '087.Irish_terrier', '088.Irish_water_spaniel', '089.Irish_wolfhound', '090.Italian_greyhound', '091.Japanese_chin', '092.Keeshond', '093.Kerry_blue_terrier', '094.Komondor', '095.Kuvasz', '096.Labrador_retriever', '097.Lakeland_terrier', '098.Leonberger', '099.Lhasa_apso', '100.Lowchen', '101.Maltese', '102.Manchester_terrier', '103.Mastiff', '104.Miniature_schnauzer', '105.Neapolitan_mastiff', '106.Newfoundland', '107.Norfolk_terrier', '108.Norwegian_buhund', '109.Norwegian_elkhound', '110.Norwegian_lundehund', '111.Norwich_terrier', '112.Nova_scotia_duck_tolling_retriever', '113.Old_english_sheepdog', '114.Otterhound', '115.Papillon', '116.Parson_russell_terrier', '117.Pekingese', '118.Pembroke_welsh_corgi', '119.Petit_basset_griffon_vendeen', '120.Pharaoh_hound', '121.Plott', '122.Pointer', '123.Pomeranian', '124.Poodle', '125.Portuguese_water_dog', '126.Saint_bernard', '127.Silky_terrier', '128.Smooth_fox_terrier', '129.Tibetan_mastiff', '130.Welsh_springer_spaniel', '131.Wirehaired_pointing_griffon', '132.Xoloitzcuintli', '133.Yorkshire_terrier']
In [20]:
import torchvision
# Get a batch of training data
inputs, classes = next(iter(loaders_scratch['train']))

for image, label in zip(inputs, classes): 
    image = image.to("cpu").clone().detach()
    image = image.numpy().squeeze()
    image = image.transpose(1,2,0)
    image = image * np.array((0.229, 0.224, 0.225)) + np.array((0.485, 0.456, 0.406))
    image = image.clip(0, 1)
     
    fig = plt.figure(figsize=(12,3))
    plt.imshow(image)
    plt.title(class_names[label])

Question 3: Describe your chosen procedure for preprocessing the data.

  • How does your code resize the images (by cropping, stretching, etc)? What size did you pick for the input tensor, and why?
  • Did you decide to augment the dataset? If so, how (through translations, flips, rotations, etc)? If not, why not?

Answer:DataLoader created for each dataset like train,test and validation Images is resized to 224 and center is cropped and added some simple data augmentation by randomly flipping and rotating the given image data. Most of the pretrained models require the input to be 224x224 images. Each color channel was normalized separately, the means are [0.485, 0.456, 0.406] and the standard deviations are [0.229, 0.224, 0.225].

(IMPLEMENTATION) Model Architecture

Create a CNN to classify dog breed. Use the template in the code cell below.

In [21]:
import torch.nn as nn
import torch.nn.functional as F

# define the CNN architecture
class Net(nn.Module):
    ### TODO: choose an architecture, and complete the class
    def __init__(self):
        super(Net, self).__init__()
        ## Define layers of a CNN
        self.conv1 = nn.Conv2d(3, 16, 3, padding=1)
        self.conv2 = nn.Conv2d(16, 32, 3, padding=1)
        self.conv3 = nn.Conv2d(32, 64, 3, padding=1)
        self.pool = nn.MaxPool2d(2, 2)
        self.fc1 = nn.Linear(64 * 28 * 28, 500)
        self.fc2 = nn.Linear(500, 133)
        self.dropout = nn.Dropout(0.25)
        self.batch_norm = nn.BatchNorm1d(num_features=500)
    
    def forward(self, x):
        ## Define forward behavior
        x = self.pool(F.relu(self.conv1(x)))
        x = self.dropout(x)
        x = self.pool(F.relu(self.conv2(x)))
        x = self.dropout(x)
        x = self.pool(F.relu(self.conv3(x)))
        x = self.dropout(x)
        x = x.view(x.size(0), -1)
        x = F.relu(self.batch_norm(self.fc1(x)))
        x = self.dropout(x)
        x = self.fc2(x)
        return x
        

#-#-# You do NOT have to modify the code below this line. #-#-#

# instantiate the CNN

model_scratch = Net()
print(model_scratch)

# move tensors to GPU if CUDA is available
if use_cuda:
    model_scratch.cuda()
Net(
  (conv1): Conv2d(3, 16, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
  (conv2): Conv2d(16, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
  (conv3): Conv2d(32, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
  (pool): MaxPool2d(kernel_size=2, stride=2, padding=0, dilation=1, ceil_mode=False)
  (fc1): Linear(in_features=50176, out_features=500, bias=True)
  (fc2): Linear(in_features=500, out_features=133, bias=True)
  (dropout): Dropout(p=0.25)
  (batch_norm): BatchNorm1d(500, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
)

Question 4: Outline the steps you took to get to your final CNN architecture and your reasoning at each step.

Answer: CNN Architecture: First layer is convolution layers has input shape of (224, 224, 3) ,pretrained models require the input to be 224x224 images. Last layer should output 133 classes.As we are training on 133 classes in image data

Convolutional layers , Maxpooling layers , usual Linear and Dropout layers used in code to avoid overfitting and produce a 133-dim output.

MaxPooling2D is slected beacuse it is good choice for classification problems.

The more convolutional layers we include, the more complex patterns in color and shape a model can detect.

New layer to have 16 filters, each with a height and width of 3. When performing the convolution, filter to jump 1 pixel at a time.

_nn.Conv2d(in_channels, out_channels, kernelsize, stride=1, padding=0)

Layer to have the same width and height as the input layer, so padding is choosen accordingly; Then, to construct this convolutional layer,code is : self.conv2 = nn.Conv2d(3, 32, 3, padding=1)

Pool layer that takes in a kernel_size and a stride after every convolution layer. This will down-sample the input's x-y dimensions, by a factor of 2

(IMPLEMENTATION) Specify Loss Function and Optimizer

Use the next code cell to specify a loss function and optimizer. Save the chosen loss function as criterion_scratch, and the optimizer as optimizer_scratch below.

In [22]:
import torch.optim as optim

### TODO: select loss function
criterion_scratch = nn.CrossEntropyLoss()

### TODO: select optimizer
optimizer_scratch = optim.SGD(model_scratch.parameters(), lr = 0.01)

(IMPLEMENTATION) Train and Validate the Model

Train and validate your model in the code cell below. Save the final model parameters at filepath 'model_scratch.pt'.

In [23]:
# the following import is required for training to be robust to truncated images
from PIL import ImageFile
ImageFile.LOAD_TRUNCATED_IMAGES = True

def train(n_epochs, loaders, model, optimizer, criterion, use_cuda, save_path):
    """returns trained model"""
    # initialize tracker for minimum validation loss
    valid_loss_min = 3.690161 #np.Inf 
    
    if os.path.exists(save_path):
        model.load_state_dict(torch.load(save_path))
    
    for epoch in range(1, n_epochs+1):
        # initialize variables to monitor training and validation loss
        train_loss = 0.0
        valid_loss = 0.0
        
        ###################
        # train the model #
        ###################
        model.train()
        for batch_idx, (data, target) in enumerate(loaders['train']):
            # move to GPU
            if use_cuda:
                data, target = data.cuda(), target.cuda()
                
            optimizer.zero_grad()
            ## find the loss and update the model parameters accordingly
            ## record the average training loss, using something like
            ## train_loss = train_loss + ((1 / (batch_idx + 1)) * (loss.data - train_loss))
            output = model(data)
            loss = criterion(output, target)
            loss.backward()
            optimizer.step()
            train_loss += loss.item()*data.size(0)
            
        ######################    
        # validate the model #
        ######################
        model.eval()
        for batch_idx, (data, target) in enumerate(loaders['valid']):
            # move to GPU
            if use_cuda:
                data, target = data.cuda(), target.cuda()
            ## update the average validation loss
            output = model(data)
            loss = criterion(output, target)
            valid_loss += loss.item()*data.size(0)
        train_loss = train_loss/len(loaders['train'].dataset)
        valid_loss = valid_loss/len(loaders['valid'].dataset)

            
        # print training/validation statistics 
        print('Epoch: {} \tTraining Loss: {:.6f} \tValidation Loss: {:.6f}'.format(
            epoch, 
            train_loss,
            valid_loss
            ))
        
        ## TODO: save the model if validation loss has decreased
        
        if valid_loss <= valid_loss_min:
            print('Validation loss decreased ({:.6f} --> {:.6f}).  Saving model ...'.format(
            valid_loss_min,
            valid_loss))
            torch.save(model.state_dict(), save_path)
            valid_loss_min = valid_loss
    # return trained model
    return model
In [24]:
#from PIL import ImageFile
#ImageFile.LOAD_TRUNCATED_IMAGES = True
# train the model
model_scratch = train(20, loaders_scratch, model_scratch, optimizer_scratch, 
                      criterion_scratch, use_cuda, 'model_scratch.pt')
Epoch: 1 	Training Loss: 1.584516 	Validation Loss: 3.691931
Epoch: 2 	Training Loss: 1.460889 	Validation Loss: 3.886051
Epoch: 3 	Training Loss: 1.340061 	Validation Loss: 3.806022
Epoch: 4 	Training Loss: 1.230662 	Validation Loss: 3.799209
Epoch: 5 	Training Loss: 1.129246 	Validation Loss: 3.938746
Epoch: 6 	Training Loss: 1.040975 	Validation Loss: 3.946384
Epoch: 7 	Training Loss: 0.938792 	Validation Loss: 3.771620
Epoch: 8 	Training Loss: 0.842769 	Validation Loss: 3.799405
Epoch: 9 	Training Loss: 0.789761 	Validation Loss: 3.886339
Epoch: 10 	Training Loss: 0.726576 	Validation Loss: 3.870545
Epoch: 11 	Training Loss: 0.665303 	Validation Loss: 3.816113
Epoch: 12 	Training Loss: 0.604955 	Validation Loss: 3.885801
Epoch: 13 	Training Loss: 0.554590 	Validation Loss: 3.859726
Epoch: 14 	Training Loss: 0.519845 	Validation Loss: 3.869554
Epoch: 15 	Training Loss: 0.491616 	Validation Loss: 3.933559
Epoch: 16 	Training Loss: 0.461225 	Validation Loss: 3.981986
Epoch: 17 	Training Loss: 0.416494 	Validation Loss: 3.913440
Epoch: 18 	Training Loss: 0.389164 	Validation Loss: 3.885455
Epoch: 19 	Training Loss: 0.372638 	Validation Loss: 3.989624
Epoch: 20 	Training Loss: 0.339920 	Validation Loss: 4.000909
In [25]:
# load the model that got the best validation accuracy
model_scratch.load_state_dict(torch.load('model_scratch.pt'))

(IMPLEMENTATION) Test the Model

Try out your model on the test dataset of dog images. Use the code cell below to calculate and print the test loss and accuracy. Ensure that your test accuracy is greater than 10%.

In [26]:
def test(loaders, model, criterion, use_cuda):

    # monitor test loss and accuracy
    test_loss = 0.
    correct = 0.
    total = 0.

    model.eval()
    for batch_idx, (data, target) in enumerate(loaders['test']):
        # move to GPU
        if use_cuda:
            data, target = data.cuda(), target.cuda()
        # forward pass: compute predicted outputs by passing inputs to the model
        output = model(data)
        # calculate the loss
        loss = criterion(output, target)
        # update average test loss 
        test_loss = test_loss + ((1 / (batch_idx + 1)) * (loss.data - test_loss))
        # convert output probabilities to predicted class
        pred = output.data.max(1, keepdim=True)[1]
        # compare predictions to true label
        correct += np.sum(np.squeeze(pred.eq(target.data.view_as(pred))).cpu().numpy())
        total += data.size(0)
            
    print('Test Loss: {:.6f}\n'.format(test_loss))

    print('\nTest Accuracy: %2d%% (%2d/%2d)' % (
        100. * correct / total, correct, total))
In [27]:
# call test function    
test(loaders_scratch, model_scratch, criterion_scratch, use_cuda)
Test Loss: 3.701931


Test Accuracy: 14% (125/836)

Step 4: Create a CNN to Classify Dog Breeds (using Transfer Learning)

You will now use transfer learning to create a CNN that can identify dog breed from images. Your CNN must attain at least 60% accuracy on the test set.

(IMPLEMENTATION) Specify Data Loaders for the Dog Dataset

Use the code cell below to write three separate data loaders for the training, validation, and test datasets of dog images (located at dogImages/train, dogImages/valid, and dogImages/test, respectively).

If you like, you are welcome to use the same data loaders from the previous step, when you created a CNN from scratch.

In [28]:
## TODO: Specify data loaders

loaders_transfer = loaders_scratch
print(loaders_transfer)
{'train': <torch.utils.data.dataloader.DataLoader object at 0x0000014D09A7B940>, 'valid': <torch.utils.data.dataloader.DataLoader object at 0x0000014D09A7BC18>, 'test': <torch.utils.data.dataloader.DataLoader object at 0x0000014D09A7BC50>}

(IMPLEMENTATION) Model Architecture

Use transfer learning to create a CNN to classify dog breed. Use the code cell below, and save your initialized model as the variable model_transfer.

In [29]:
import torchvision.models as models
import torch.nn as nn

## TODO: Specify model architecture 
model_transfer = models.resnet50(pretrained=True)


if use_cuda:
    model_transfer = model_transfer.cuda()

Question 5: Outline the steps you took to get to your final CNN architecture and your reasoning at each step. Describe why you think the architecture is suitable for the current problem.

Answer: Convolutional neural networks Several pre-trained models used in transfer learning are based on large convolutional neural networks (CNN).In general, CNN was shown to excel in a wide range of computer vision task

for our dog classifier application we are selected resnet50 trained on ImageNet available from torchvision. ResNet is a powerful backbone model that is used very frequently in many computer vision tasks. ResNet uses skip connection to add the output from an earlier layer to a later layer. This helps it mitigate the vanishing gradient problem. The ResNet-50 model consists of 5 stages each with a convolution and Identity block. Each convolution block has 3 convolution layers and each identity block also has 3 convolution layers. The ResNet-50 has over 23 million trainable parameters.

The classifier part of the model is a single fully-connected layer:

(fc): Linear(in_features=2048, out_features=1000, bias=True) we need to replace the classifier (133 classes), but the features will work perfectly on their own.

In [30]:
# Freeze parameters so we don't backprop through them
for param in model_transfer.parameters():
    param.requires_grad = False
# Replace the last fully connected layer with a Linnear layer with 133 out features
model_transfer.fc = nn.Linear(2048, 133)
if use_cuda:
    model_transfer = model_transfer.cuda()

(IMPLEMENTATION) Specify Loss Function and Optimizer

Use the next code cell to specify a loss function and optimizer. Save the chosen loss function as criterion_transfer, and the optimizer as optimizer_transfer below.

In [31]:
criterion_transfer = nn.CrossEntropyLoss()
optimizer_transfer = optim.Adam(model_transfer.fc.parameters(), lr=0.001)

(IMPLEMENTATION) Train and Validate the Model

Train and validate your model in the code cell below. Save the final model parameters at filepath 'model_transfer.pt'.

In [32]:
# train the model
n_epochs = 15
model_transfer = train(n_epochs, loaders_transfer, model_transfer, optimizer_transfer, criterion_transfer, use_cuda, 'model_transfer.pt')

# load the model that got the best validation accuracy (uncomment the line below)
#model_transfer.load_state_dict(torch.load('model_transfer.pt'))
Epoch: 1 	Training Loss: 0.375422 	Validation Loss: 0.667688
Validation loss decreased (3.690161 --> 0.667688).  Saving model ...
Epoch: 2 	Training Loss: 0.328169 	Validation Loss: 0.719755
Epoch: 3 	Training Loss: 0.308754 	Validation Loss: 0.691226
Epoch: 4 	Training Loss: 0.292466 	Validation Loss: 0.670980
Epoch: 5 	Training Loss: 0.310260 	Validation Loss: 0.623496
Validation loss decreased (0.667688 --> 0.623496).  Saving model ...
Epoch: 6 	Training Loss: 0.302327 	Validation Loss: 0.788290
Epoch: 7 	Training Loss: 0.281854 	Validation Loss: 0.660981
Epoch: 8 	Training Loss: 0.276763 	Validation Loss: 0.692088
Epoch: 9 	Training Loss: 0.303940 	Validation Loss: 0.677363
Epoch: 10 	Training Loss: 0.255775 	Validation Loss: 0.766288
Epoch: 11 	Training Loss: 0.279244 	Validation Loss: 0.709858
Epoch: 12 	Training Loss: 0.269571 	Validation Loss: 0.708402
Epoch: 13 	Training Loss: 0.271889 	Validation Loss: 0.833093
Epoch: 14 	Training Loss: 0.253956 	Validation Loss: 0.702646
Epoch: 15 	Training Loss: 0.246877 	Validation Loss: 0.737150
In [33]:
model_transfer.load_state_dict(torch.load('model_transfer.pt'))

(IMPLEMENTATION) Test the Model

Try out your model on the test dataset of dog images. Use the code cell below to calculate and print the test loss and accuracy. Ensure that your test accuracy is greater than 60%.

In [34]:
test(loaders_transfer, model_transfer, criterion_transfer, use_cuda)
Test Loss: 0.748349


Test Accuracy: 82% (690/836)

(IMPLEMENTATION) Predict Dog Breed with the Model

Write a function that takes an image path as input and returns the dog breed (Affenpinscher, Afghan hound, etc) that is predicted by your model.

In [36]:
### TODO: Write a function that takes a path to an image as input
### and returns the dog breed that is predicted by the model.

# list of class names by index, i.e. a name can be accessed like class_names[0]
class_names = [item[4:].replace("_", " ") for item in image_datasets['train'].classes]

def predict_breed_transfer(img_path):
    # load the image and return the predicted breed
    image_tensor = image_to_tensor(img_path)
    
    
    if use_cuda:
        image_tensor = image_tensor.cuda()
    output = model_transfer(image_tensor)
        # convert output probabilities to predicted class
    _, preds_tensor = torch.max(output, 1)
    pred = np.squeeze(preds_tensor.numpy()) if not use_cuda else np.squeeze(preds_tensor.cpu().numpy())

    return class_names[pred]
In [37]:
def display_image(img_path, title="Title"):
    image = Image.open(img_path)
    plt.title(title)
    plt.imshow(image)
    plt.show()
In [38]:
import random

# Try out the function
for image in random.sample(list(human_files_short), 4): 
    predicted_breed = predict_breed_transfer(image)
    display_image(image, title=f"Predicted:{predicted_breed}")

Step 5: Write your Algorithm

Write an algorithm that accepts a file path to an image and first determines whether the image contains a human, dog, or neither. Then,

  • if a dog is detected in the image, return the predicted breed.
  • if a human is detected in the image, return the resembling dog breed.
  • if neither is detected in the image, provide output that indicates an error.

You are welcome to write your own functions for detecting humans and dogs in images, but feel free to use the face_detector and dog_detector functions developed above. You are required to use your CNN from Step 4 to predict dog breed.

Some sample output for our algorithm is provided below, but feel free to design your own user experience!

Sample Human Output

(IMPLEMENTATION) Write your Algorithm

In [51]:
### TODO: Write your algorithm.
### Feel free to use as many code cells as needed.

def run_app(img_path):
    # check if image has juman faces:
    if (face_detector(img_path)):
        print("Hello Human!")
        predicted_breed = predict_breed_transfer(img_path)
        display_image(img_path, title=f"Predicted:{predicted_breed}")
        
        print("You look like a ...")
        print(predicted_breed.upper())
    # check if image has dogs:
    elif dog_detector(img_path):
        print("Hello Dog!")
        predicted_breed = predict_breed_transfer(img_path)
        display_image(img_path, title=f"Predicted:{predicted_breed}")
        
        print("Your predicted breed is ...")
        print(predicted_breed.upper())
    else:
        print("Oh, we're sorry! We couldn't detect any dog or human face in the image.")
        display_image(img_path, title="...")
        print("Try another!")
    print("\n")

Step 6: Test Your Algorithm

In this section, you will take your new algorithm for a spin! What kind of dog does the algorithm think that you look like? If you have a dog, does it predict your dog's breed accurately? If you have a cat, does it mistakenly think that your cat is a dog?

(IMPLEMENTATION) Test Your Algorithm on Sample Images!

Test your algorithm at least six images on your computer. Feel free to use any images you like. Use at least two human and two dog images.

Question 6: Is the output better than you expected :) ? Or worse :( ? Provide at least three possible points of improvement for your algorithm.

Answer: (Three possible points for improvement)

Some improvements: 1.Hyper-parameter tunings: weight initializings, learning rates, drop-outs, batch_sizes, and optimizers will be helpful to improve performances. 2.More image datasets of dogs will improve training models. Also, more image augmentations trials (flipping vertically, move left or right, etc.) will improve performance on test data.

3.Increase the accuracy of the dog_breed_predictor, by increasing the number of training episodes and maybe using a better feature detector, e.g. the one from the InceptionV3 network.

In [52]:
## TODO: Execute your algorithm from Step 6 on
## at least 6 images on your computer.
## Feel free to use as many code cells as needed.

## suggested code, below
for file in np.hstack((human_files[:3], dog_files[:3])):
    run_app(file)
Hello Human!
You look like a ...
FRENCH BULLDOG


Hello Human!
You look like a ...
AMERICAN WATER SPANIEL


Oh, we're sorry! We couldn't detect any dog or human face in the image.
Try another!


Hello Dog!
Your predicted breed is ...
AFFENPINSCHER


Hello Dog!
Your predicted breed is ...
AFFENPINSCHER


Hello Dog!
Your predicted breed is ...
AFFENPINSCHER


In [53]:
run_app("test1.jpg")
Hello Human!
You look like a ...
IRISH WOLFHOUND


In [54]:
run_app("test3.jpg")
Oh, we're sorry! We couldn't detect any dog or human face in the image.
Try another!


In [55]:
run_app("test5.jpg")
Hello Human!
You look like a ...
KERRY BLUE TERRIER


In [56]:
run_app("test6.jpg")
Hello Human!
You look like a ...
NEWFOUNDLAND


In [57]:
run_app("test7.jpg")
Oh, we're sorry! We couldn't detect any dog or human face in the image.
Try another!


In [58]:
run_app("test8.jpg")
Oh, we're sorry! We couldn't detect any dog or human face in the image.
Try another!


In [59]:
run_app("test9.jpg")
Hello Dog!
Your predicted breed is ...
LABRADOR RETRIEVER


In [60]:
run_app("test10.jpg")
Hello Dog!
Your predicted breed is ...
LABRADOR RETRIEVER


In [ ]:
 
In [ ]: